© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 27629
    Prev
    Next
    INDEX
    Explanations

    references related to technical specifications and software details

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    376
    +0.12
    0.7%
    495
    +0.11
    0.6%
    141
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    12
    +0.12
    0.02
    495
    +0.11
    -0.01
    89
    +0.11
    0.06
    Negative Logits
    Ĥ
    -1.80
    Ĩ
    -1.65
    ª
    -1.64
    Ĵ
    -1.54
    Ļª
    -1.50
     melan
    -1.46
    thirds
    -1.46
    ¦
    -1.45
    ¾
    -1.44
    ĻĤ
    -1.44
    POSITIVE LOGITS
    ETHOD
    1.71
     injection
    1.64
    urger
    1.47
    urg
    1.40
    ault
    1.40
    CTL
    1.37
     reversed
    1.32
    zzles
    1.32
     injected
    1.29
     ARISING
    1.27
    Activations Density 9.926%

    No Known Activations